TARBP2 Suppresses Ubiquitin-Proteasomal Degradation of HIF-1α in Breast Cancer

TAR (HIV-1) RNA binding protein 2 (TARBP2) is an RNA-binding protein participating in cytoplasmic microRNA processing. Emerging evidence has shown the oncogenic role of TARBP2 in promoting cancer progression, making it an unfavorable prognosis marker for breast cancer. Hypoxia is a hallmark of the tumor microenvironment which induces hypoxia-inducible factor-1α (HIF-1α) for transcriptional regulation. HIF-1α is prone to be rapidly destabilized by the ubiquitination–proteasomal degradation system. In this study, we found that TARBP2 expression is significantly correlated with induced hypoxia signatures in human breast cancer tissues. At a cellular level, HIF-1α protein level was maintained by TARBP2 under either normoxia or hypoxia. Mechanistically, TARBP2 enhanced HIF-1α protein stability through preventing its proteasomal degradation. In addition, downregulation of multiple E3 ligases targeting HIF-1α (VHL, FBXW7, TRAF6) and reduced ubiquitination of HIF-1α were also induced by TARBP2. In support of our clinical findings that TARBP2 is correlated with tumor hypoxia, our IHC staining showed the positive correlation between HIF-1α and TARBP2 in human breast cancer tissues. Taken together, this study indicates the regulatory role of TARBP2 in the ubiquitination–proteasomal degradation of HIF-1α protein in breast cancer.


Introduction
Hypoxia-inducible factor-1α (HIF-1α) is a versatile transcriptional factor that exerts its transactivation ability for transcriptional regulation [1]. Utilizing its transcriptional activity, HIF-1α affects several cancer properties, including tumorigenesis, apoptosis, genomic instability, metastasis, and invasion, and is therefore important in cancer progression [2]. Involved in numerous cellular processes, HIF-1α is tightly regulated by oxygen concentration. Under normoxic conditions, HIF-1α acts as an oxygen-labile protein with a short half-life, which means that as soon as HIF-1α is synthesized it gets degraded via the proteasomal or lysosomal degradation pathways [3]. HIF-1α is hydroxylated by prolyl-4-hydroxylases (PHDs) in the oxygen-dependent degradation domain (ODDD) at the proline residues (P402 and P564), which is recognized by its E3 ligase, von Hippel-Lindau protein (VHL), followed by ubiquitination, thereby leading to its proteasomal degradation. During hypoxic conditions, PHDs are inhibited and HIF-1α escapes the fate of degradation and becomes dominant to exert its effect. In addition to hypoxia, deubiquitination of HIF-1α is an alternate pathway which inhibits degradation, and this process is facilitated by a group of deubiquitinating enzymes (DUBs). These deubiquitinases (DUBs) function via removing the ubiquitinated marks or editing the ubiquitin chains in the target proteins to recycle ubiquitin for maintaining ubiquitin homeostasis [4]. Consequently, protein degradation controls the main regulation of HIF-1α, which results from the fluctuations in ubiquitination and deubiquitination. Although several E3 ligases and DUBs have been individually reported to affect HIF-1α expression [4], the global regulation of these two processes is less well understood. TAR (HIV-1) RNA binding protein 2 (TARBP2) is a multifunctional RNAbinding protein known to be involved in microRNA (miRNA) biogenesis, viral infection, and tumorigenesis [5]. TARBP2 is one of the miRNA biogenesis factors, participating in the cytoplasmic processing of miRNA and acting as a cofactor with Dicer [6]. TARBP2 also regulates HIV-1 gene expression through interacting with TAR [6]. Goodarzi et al. reported that TARBP2 is overexpressed in metastatic tumor cells and promotes colony formation and invasion ability through destabilizing two metastasis suppressors, including amyloid precursor protein (APP) and ZNF395 [7]. In addition, TARBP2 is reported to promote tumor-induced angiogenesis through degradation of mRNAs coding for antiangiogenetic factors, including thrombospondin1/2 (THBS1/2), tissue inhibitor of metalloproteinases 1 (TIMP1), and serpin family F member 1 (SERPINF1) [8]. TARBP2 has been shown to enhance a transformed phenotype and tumorigenesis in vivo and is considered an unfavorable prognostic marker for breast cancer [9,10]. These studies suggest an oncogenic role of TARBP2 in cancer, which might be involved in numerous unknown mechanisms. Although HIF-1α and TARBP2 are both crucial for cancer development, the relationship between these two proteins remains to be clarified. In this study, we clarify the relationship between TARBP2 and HIF-1α and show that TARBP2 positively regulates HIF-1α by inhibiting its poly-ubiquitination and proteasomal degradation.

Western Blot
After cells reached 70-80% confluence, the whole cell lysates were collected in NETN buffer (20 mM Tris-Base, pH 8.0, 150 mM NaCl, 1 mM EDTA, 0.5% Nonidet P-40) combined with protease inhibitors and phosphatase inhibitors. The proteins were purified by centrifugation at 14,000 rpm, 4 • C and quantified using Dual-Range TM Bradford reagent. Then, 60-70 µg of proteins were fractionated on the 8-12% Tris-glycine gel and electrophoretically transferred onto PVDF membranes according to the manufacture's protocols (Bio-Rad). After protein transfer, the membranes were blocked with 5% non-fat milk in TBST buffer (20 mM Tris base, 150 mM NaCl, and 0.1% Tween 20) for at least 60 min, and then washed in TBST for 10 min. Membranes were incubated with primary antibodies diluted in 2% BSA in TBST at 4 • C overnight and incubated with anti-mouse or anti-rabbit secondary antibodies for 60 min having been washed with TBST three times for 10 min on the secondary day. Protein expressions were visualized using an ECL detection system according to the manufacture's protocols.

RNA Extraction and Reverse Transcriptase Real-Time PCR
Total RNA was extracted by Trizol reagent according to the manufacturer's instructions and quantified with the NanoDrop Lite spectrophotometer. Complementary DNA (cDNA) was synthesized with 100-200 ng of RNA in an MJ Mini thermal cycler. For reverse transcription, reaction mixtures containing template RNA, random hexamer, and dNTP were incubated at 65 • C for 5 min, following which, reaction buffer, RNase inhibitor, RevertAid RT and nuclease-free water were added for incubation at 25 • C for 5 min and 42 • C for 60 min, terminating at 70 • C for 5 min. The amplifications of cDNA were performed using the SYBR Green PCR Master Mix in an Applied Biosystem Step One Real-time PCR system (Applied Biosystems) according to the manufacturer's protocols. The reaction mixture, including Fast SYBR™ Green PCR Master Mix, forward primer, reverse primer, nuclease-free water and cDNA templates, was incubated at 95 • C for 20 s, 40 cycles of 95 • C for 3 s and 60 • C for 30 s. Each sample was carried out in duplicate and GAPDH was used to normalize the target genes.

Tissue Microarray
The tissue array was obtained from Taipei Medical University Joint Biobank and purchased from Biomax US, lnc. (Derwood, MD, USA) Waivers of informed consent were approved by A-ER-106-483 from NCKU Hospital. The study was conducted according to ethical approval by the Ethics Committee of Taipei Medical University (approval numbers N202001052 and N202103065).

Immunohistochemistry
Breast cancer tissue array was used for immunohistochemistry (IHC) staining. The TARBP2 and HIF-1α antigen retrieval program was incubated in boiling citrate buffer (pH 6.0) for 20 min. Blocking buffer (TA00C2, BioTnA, Kaohsiung, Taiwan) and H 2 O 2 were used to block endogenous peroxidase of the tissue for 60 min each. Tissue sections were stained with primary antibodies, followed by HRP-conjugated secondary antibodies, to observe TARBP2 (code No. 3-2, customized antibody, Leadgene Biomedical, Tainan, Taiwan) and HIF-1α (BS3514, Bioworld Technology, Inc., St. Louis, MO, USA) at 1:3000 and 1:100 dilution factors. Primary antibodies of TARBP2 and HIF-1α were incubated overnight at 4°C and room temperature, respectively, and expression levels were measured using the TAlink mouse/rabbit polymer detection system (TAHC04D, BioTnA, Kaohsiung, Taiwan).

Statistical Analysis
All experiments were carried out at least three times and results were calculated as the means ± standard error of mean (SEM). The data was analyzed by t-tests for two individual groups or two-way ANOVA for groups with two variables in Prism7 software. Values between different groups were considered significant when p was less than 0.05.

TARBP2 Upregulates HIF-1α Expression
To investigate the association between TARBP2 and hypoxia, we utilized the cohort of breast cancer patients from the Cancer Genome Atlas (TCGA) and analyzed hypoxia scores, which were calculated using the mRNA-based signatures developed by Buffa and Winter [11,12]. TARBP2 was found to be positively correlated with both Winter (Pearson correlation coefficient r = 0.2146, p < 0.0001) and Buffa (Pearson correlation coefficient r = 0.1539, p < 0.0001) hypoxia scores, which indicated that TARBP2 exhibits a crucial role in hypoxia ( Figure 1A). Since HIF-1α is a master transcription factor that is dominant in hypoxia [13], we investigated the expression of HIF-1α in TARBP2-overexpressing cells. Increased HIF-1α expression was observed following TARBP2 overexpression in MCF-7 and MDA-MB-231 cells ( Figure 1B). Additionally, decreased HIF-1α expression was observed in TARBP2 knockdown MCF-7 cells, suggesting that TARBP2 induces HIF-1α expression ( Figure 1C). Moreover, elevated HIF-1α expression in hypoxia was found to be abrogated by knockdown of TARBP2 ( Figure 1D). These data indicated that TARBP2 plays an important role in upregulating HIF-1α expression under both normoxia and hypoxia. Since TARBP2 is a key factor in cytoplasmic miRNA biogenesis and the C4 domain is required for Dicer-mediated miRNA processing [14], we used the C4 domain-truncated TARBP2 (TARBP2 ∆C4) to study whether its function in miRNA biogenesis is required for HIF-1α regulation. The results showed that increased HIF-1α expression is still observed in TARBP2 ∆C4-overexpressing cells as TARBP2 full length, indicating that TARBP2 acts through an miRNA biogenesis-independent pathway to upregulate HIF-1α ( Figure 1E).

TARBP2 Inhibits Proteasomal Degradation of HIF-1α
Since our results showed that TARBP2 promotes HIF-1α protein expression, we investigated whether the mRNA expression of HIF-1α is also affected. Unchanged expression of HIF-1α mRNA was observed in TARBP2-overexpressing cells; however, the downstream expression of HIF-1α, including CA9, EGLN1, PLOD3, EFNA1, and ALDOA, increased (Figure 2A). These results suggest that TARBP2 enhances HIF-1α protein expression as well as its transcriptional activity. To further elucidate the mechanism of TARBP2-mediated upregulation of HIF-1α, we used cycloheximide (CHX) to block its protein synthesis. In TARBP2-overexpressing cells, enhanced protein stability of HIF-1α was observed ( Figure 2B). Since the basal level of HIF-1α is relatively low under normoxia, we also treated cells with the hypoxia-mimetic agent CoCl 2 to enhance HIF-1α accumulation ( Figure 2C). In support of our results showing TARBP2-induced HIF-1α under both hypoxia and normoxia, HIF-1α exhibited reduced protein stability when TARBP2 expression was knocked down in both DMSO-and CoCl 2 -treated cells ( Figure 2C). HIF-1α is an unstable protein that is rapidly degraded via both proteasomal and lysosomal pathways [13]. Since TARBP2 upregulates HIF-1α through promoting its protein stability, we investigated the underlying pathway using proteasome inhibitor MG132 and lysosome inhibitor NH 4 Cl/ chloroquine (CQ) treatment. In TARBP2 knockdown cells, HIF-1α was found to be downregulated and this phenomenon was abrogated upon MG132 treatment; however, HIF-1α expression remained decreased upon NH 4 Cl/CQ treatment ( Figure 2D). These data indicated that TARBP2 upregulates HIF-1α expression through inhibiting its proteasomal degradation, thereby leading to enhanced protein stability and transcriptional activity.

TARBP2 Is Positively Correlated with HIF-1α in Breast Cancer Tissues
Having discovered that TARBP2 upregulates HIF-1α through inhibiting its proteasomal degradation, we investigated the correlation between TARBP2 and HIF-1α in patients with breast cancer. The protein expression of TARBP2 and HIF-1α was determined by IHC staining using a tissue microarray of human breast cancer. Higher HIF-1α (HIF-1α high ) expression was observed in tumor tissues with higher TARBP2 (TARBP2 high ) levels ( Figure 6A, upper panel), and vice versa ( Figure 6A, bottom panel). The statistical correlation between the IHC scores of TARBP2 and HIF-1α were analyzed ( Figure 6B). Compared to the TARBP2 low group, the percentage of HIF-1α high was elevated from 38.3% to 61.7% in the TARBP2 high group, with decreased HIF-1α high percentage from 68.2% to 31.2% ( Figure 6B).

Conclusions
These data indicated that TARBP2 protein expression is positively correlated with HIF-1α in human breast cancer. Taken together, we revealed that TARBP2 is a HIF-1α positive regulator in both normoxic and hypoxic conditions. Mechanistically, TARBP2 downregulates multiple E3 ligases, reduces HIF-1α ubiquitination and proteasomal degradation of HIF-1α, leading to elevated HIF-1α protein accumulation and transcriptional activation ( Figure 6C).

Discussion
HIF-1α is an important protein that has been shown to play an important role in cancer development. Exhibiting a short half-life, HIF-1α is degraded via the proteasome degradation pathway [16]. The regulation of HIF-1α has been reported previously in several studies focusing on its E3 ligases and deubiquitinases responsible for promoting or inhibiting the degradation pathway [4]. VHL, a well-studied E3 ligase of HIF-1α, recognizes ubiquitin-marked HIF-1α and drives its degradation under normoxia [17]. On the other hand, MDM2 is an E3 ligase of HIF-1α that forms a complex with p53 and leads to HIF-1α degradation in a p53-dependent manner under hypoxia [18]. Ubiquitination is a dynamic process that requires a series of enzymes (E1, E2, and E3) for conjugating ubiquitin to the target, and deubiquitinases (DUBs) remove the ubiquitin and maintain homeostasis [15]. USP20 is a DUB that counteracts VHL-mediated ubiquitination of HIF-1α [19]. USP8 and UCHL1 have been reported to remove ubiquitin from HIF-1α and HIF-2α, marked by VHL, to inhibit degradation [20]. MCPIP1 deubiquitylates HIF-1α to rescue its degradation [21]. These studies have reported the impact of a single E3 ligase or DUB on HIF-1α. In our study, we found that TARBP2 downregulates E3 ligases globally to attenuate the ubiquitination of HIF-1α, thereby resulting in enhanced expression of HIF-1α. These findings suggest that TARBP2 acts as a major modulator of HIF-1α ubiquitination status through controlling the expression of HIF-1α-related E3 ligases. Featured as an RNA-binding protein, TARBP2 is known to be a microRNA biogenesis factor involved in the cytoplasmic maturation of miRNAs [22][23][24]. Lin et al. found that the expression of TARBP2 is positively correlated with the stage in breast cancer, indicating that TARBP2 is an unfavorable prognostic marker [9]. The oncogenic effect of TARBP2 has also been reported in metastatic breast cancer indicating that the overexpression of TARBP2 destabilizes amyloid precursor proteins (APPs) and ZNF395 transcripts containing TARBP2-binding structural elements (TBSEs), thereby leading to the degradation of metastasis-suppressor transcripts associated with Alzheimer's and Huntington's disease, respectively [7]. On the other hand, TARBP2 is reported to be transcriptionally repressed in a PHD-dependent manner under hypoxia [25]. Let-7f-5p has also been found to be induced under hypoxia to suppress TARBP2 [26]. These studies indicate another layer of negative regulation from hypoxia to TARBP2. In our findings, we suggest a novel function of TARBP2 in regulating HIF-1α protein turnover, which transcriptionally downregulates HIF-1α-related E3 ligases. In TARBP2 high tumors, HIF-1α expression was found to be higher than that in TARBP2 low tumors, thereby suggesting that TARBP2 may serve as a crucial protein regulator of HIF-1α and exhibit an oncogenic role in cancer progression.  Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board (or Ethics Committee) of National Cheng Kung University Hospital (Tainan, Taiwan) (IRB A−ER−106−483) and Taipei Medical University (approval numbers N202001052 and N202103065).

Informed Consent Statement:
The tissue array was obtained from Taipei Medical University Joint Biobank and Biomax US, lnc. Waivers of informed consent are approved and acquirement of patient samples are all followed by National Cheng Kung University Hospital IRB and Ethics Committee of Taipei Medical University. Data Availability Statement: All images and raw data available on request. All other data are available from the corresponding authors.

Conflicts of Interest:
The authors declare that they have no competing interests.